Can Human Brain Connectivity explain Verbal Working Memory?

Abstract

Abstract Introduction: Humans are able to learn and use a broad range of words and other symbols, whereas Monkeys are limited to acquiring small vocabularies of signs, including sounds and gestures. Although evolutionary changes on network architecture and connectivity features within the left-perisylvian regions has been reported, their functional contribution on symbol formation and verbal working memory are poorly understood. Methods: Here, we used brain-constrained neural network of the frontotemporal and occipital cortices mimicking key neuroanatomical distinctions between human and non-human primates. Results: Our comparative analysis of models shows that the human model, characterized by denser inter-area connectivity, gives rise to larger cell assemblies with distinct semantic-specific topography compared to the less densely connected monkey models. Additionally, by simulating auditory word recognition, we observed the emergence of verbal working memory by longer neural reverberation activity in the human architecture compared to those of monkeys. Interestingly, these observations are consistent across different model types, including a basic meanfield and a spiking neural model. Conclusions: These findings shed light on the structural underpinnings of human-specific verbal working memory, a crucial feature for the acquisition of an expansive vocabulary.